Context Attention deficit hyperactivity disorder (ADHD) is associated with widespread cognitive impairments, but it is not known whether the apparent multiple impairments share etiological roots, or whether separate etiological pathways exist. A better understanding of the etiological pathways is important for the development of targeted interventions and for identification of suitable intermediate phenotypes for molecular genetic investigations. Objective To determine, using a multivariate familial factor analysis approach, whether one or more familial factors underlie the slow and variable reaction times (RTs), impaired response inhibition, sustained attention, and choice impulsivity that are associated with ADHD. Design An ADHD and control sibling-pair design. Setting Belgium, Germany, Ireland, Israel, Spain, Switzerland and the United Kingdom. Participants The sample consisted of 1265 participants, aged 6 to 18 years: 464 probands with ADHD and 456 of their siblings (524 with ADHD combined subtype), and 345 control participants. Main Outcome Measures Performance on a four-choice RT task, a go/no-go inhibition task and a choice-delay task. Results The final model consisted of two familial factors. The larger factor, reflecting 85% of the familial variance of ADHD, captured 98-100% of the familial influences on mean RT and RT variability. The second smaller factor, reflecting 12.5% of the familial variance of ADHD, captured 62-82% of the familial influences on commission and omission errors on the go/no-go task. Choice impulsivity was excluded in the final model, due to poor fit. Conclusions The findings suggest the existence of two familial pathways to cognitive impairments in ADHD and indicate promising cognitive targets for future molecular genetic investigations. The familial distinction between the two cognitive impairments is consistent with recent theoretical models – a developmental model and an arousal-attention model – on two separable underlying processes in ADHD. Future research that tests the familial model within a developmental framework may inform developmentally-sensitive interventions.